<p>Tunnel boring machines (TBMs) are frequently used in tunnel excavation due to its advantages of high efficiency and environmental protection. No existing study provides quantitative characterization of rock-machine interaction in tunnel widening using a TBM. This study develops a novel tunnelling test platform, named DGTBM-A, to simulate the TBM tunnelling process. It aims to quantitatively analyze the characteristics of rock-machine interaction in tunnel widening. The platform can simulate the tunnelling operations and rock conditions to reflect the rock-TBM responses. The components and functions of the platform are illustrated. We prepared six cubic sandstone specimens with varying strengths and centrally prepped boreholes of different diameters. The influences of thrust, borehole diameter, and rock strength on the rock-TBM interaction responses are analyzed quantitatively using both performance indices and rock muck information. The performance indices include advance rate, penetration rate, specific energy, and field penetration index. The rock muck information includes chip size and shape. It shows that higher thrusts improve the tunnelling efficiency but increase the specific energy consumption. Borehole with larger diameter enhances the TBM tunnelling efficiency without significantly raising energy costs, though the required torque rise substantially. Rocks with higher strength present greater excavation challenges, with lower advance rate and higher required torque. Rock muck analysis indicates that larger thrusts and boreholes break up more effective rock breaking, producing a higher proportion of fine chips. These findings highlight the complex interplay between TBM operating parameters and rock conditions, providing insights for optimizing TBM performance in tunnel widening projects.</p>